基于行为博弈进化算法的并联混合动力汽车控制策略参数优化

Parameters Optimization of PHEV Control Strategy with Game Evolutionary Algorithm Based on Behavioral Game Theory

控制策略是影响并联混合动力汽车(PHEV)性能的重要组成部分。针对混合动力汽车控制策略优化问题,设计了基于行为博弈进化算法的并联混合动力汽车控制策略参数优化方法(PGEA);该方法以行为博弈进化算法为优化工具,以混合动力汽车的加速性能、最大爬坡度、最高车速等动力性能指标为约束条件,采用ADVISOR仿真并联混合动力汽车;以最小化燃油消耗与污染物排放总量为优化目标。30组仿真实验结果表明:(1)优化后的系统百公里燃油消耗与污染物排放之和最大降幅为23.68%,最小降幅为18.72%,平均下降22.13%;(2)发动机效率、电动机效率和系统整体效率至少分别提高了16.67%、44.44%和18.39%;(3)PGEA比基于Pareto最优原理的混合动力汽车多目标优化进化算法能够获得精度更高的解。

The control strategy is a significant part of a parallel hybrid electric vehicle( PHEV) affecting its performance. Aimed at the control strategy optimization problem,the parameters optimization of PHEV control strategy with game evolutionary algorithm base was investigated on behavioral game theory( PGEA). The optimal goal is to minimize the sum of consumption and pollutant emissions under the constraint conditions of the PHEV's acceleration performance,maximum gradability,and top speed. PGEA employs game evolutionary algorithm base on behavioral game theory to implement the evolutionary process and uses ADVISOR to simulate PHEV to evaluate the candidate solutions. The simulation results show that compared with the original system,the sum of fuel consumption and emissions per 100 km dropped by a maximum of 23. 68,a minimum of 18. 72 and an average of22. 13%. The electric motor efficiency,the combustion engine efficiency,and the overall system efficiency increased by at least of 16. 67%,44. 44%,and 18. 39%,respectively. The proposed PGEA outperforms the compared algorithm of multi-objective optimization of hybrid electrical vehicle based on Pareto optimality on the accuracy.